TRANSACTIONS OF THE AMERICAN CLINICAL AND CLIMATOLOGICAL ASSOCIATION, VOL.

120, 2009

DO THE OBESE HAVE LOWER BODY TEMPERATURES?

A NEW LOOK AT A FORGOTTEN VARIABLE
IN ENERGY BALANCE

LEWIS LANDSBERG and (by invitation) JAMES B. YOUNG,

WILLIAM R. LEONARD, ROBERT A. LINSENMEIER, FRED W. TUREK

CHICAGO, IL

ABSTRACT
Understanding the pathogenesis of obesity is now more important than
ever, given the remarkable world-wide epidemic. This paper explores the
potential role of core temperature in energy balance, and develops the
hypothesis that basal temperature and changes in the temperature re-
sponse in various situations contribute to the enhanced metabolic effi-
ciency of the obese state. The argument is based on the important contri-
bution that heat production makes in establishing the basal or resting
metabolic rate, as well as on an analysis of the adaptive role played by
changes in temperature in response to environmental challenge. If this
hypothesis is validated, new therapeutic approaches may ensue.

Total Energy Expenditure and the Energy Balance Equation

In sedentary humans, estimates suggest that physical activity ac-
counts for about 10% of energy expenditure, while basal or resting
metabolic rate (RMR) accounts for about 80%, and adaptive thermo-
genesis rounds out the remaining 10%. The latter refers to sympathet-
ically mediated heat production in response to cold or dietary intake.
RMR, in turn, has two major components: “essential” heat, which
constitutes about 1/3rd of RMR, represents the energy required to
maintain cellular integrity and the ionic environment; the second
component of RMR is heat production for homeothermy which
amounts to about 2/3rds of the metabolic rate (1–3). Although these
estimates are rough and may underestimate the contribution of phys-
ical activity depending on the definition of “sedentary,” overall they
imply that up to 50% of energy expenditure in sedentary man is
dedicated to maintaining a constant core temperature. This extraordi-
Correspondence and reprint requests: Lewis Landsberg, Northwestern University Compre-
hensive Center on Obesity, 750 N. Lake Shore Drive, Rubloff Suite 9-976, Chicago, IL 60611,
Tel: 312-908-1892, Fax: 312-503-6743, E-mail: l-landsberg@northwestern.edu
Potential Conflicts of Interest: None disclosed.
287
288 LEWIS LANDSBERG ET AL

nary observation is validated by the fact that the metabolic rate of a

mouse is many-fold higher than that of a lizard of the same weight (3).
What does this mean in terms of energy balance? Consider the
energy balance equation:

Energy Intake ⫽ Energy Output ⫹ Storage

When intake is equal to output, the individual is in energy balance,

and there is no change in storage (fat). Increases in any of the compo-
nents of energy expenditure would expand the range of intakes over
which energy balance could be maintained. Conversely, decreases in
any of the output components would reduce the range of intakes over
which balance could be achieved.
Can this simple tautological expression help to understand the current
epidemic of obesity and its related disorders? Does it help to understand
why humans are heir to obesity? Is there more to obesity than “gluttony
and sloth”? And if so, what is the basis of the differences between lean
and obese?

Metabolic Efficiency
In this context, metabolic efficiency refers to individual differences in
the dissipation of calories as heat. In a classic overfeeding experiment
involving twin pairs, a three-fold variation in weight gain was noted
among the participants who consumed an additional 1000 kcal/day, 6
days per week, for three months, while maintaining a sedentary life-style
and limiting activity (4). Some of the subjects dissipated almost no por-
tion of the increased calories while others dissipated as much as 60%.
The differences were greater between twin pairs than within twin pairs,
implying an inherited disposition to metabolic efficiency.
Those who dissipate excess calories may be said to have a less
efficient metabolism than those who store ingested calories as fat with
less dissipated as heat. Those with the more efficient metabolism resist
famine better, but have a predisposition to become obese when faced
with an abundant food supply and in the presence of dietary excess.
Given the ever increasing prevalence of obesity in the US and the
world, understanding the basis of these differences in metabolism is of
critical importance (5).

Compartments of Energy Output

The three major compartments of energy expenditure include: phys-
ical activity, adaptive or facultative thermogenesis and basal or resting
metabolic rate (RMR).
 DO THE OBESE HAVE LOWER TEMPERATURES? 289

Physical Activity. In sedentary man, purposeful and non-pur-

poseful movement account for about 10% of total energy output (2). In
people who are physically active, the activity component of energy
output is, of course, correspondingly greater. In the study of metabolic
efficiency described above, differences in activity were unlikely causes
of the differences in weight gain, pointing to differences in metabolic
efficiency as the cause of differences in weight gain.
Adaptive (Facultative) Thermogenesis. This category also ac-
counts for about 10% of total energy expenditure. Heat is generated by
chemical reactions that address a physiologic need, most notably cold
exposure (non-shivering thermogenesis) or dietary intake (diet-in-
duced thermogenesis). The processes that generate heat in these sit-
uations are regulated by the sympathetic nervous system. Thyroid
hormones play a permissive role (6). In rodents and small mammals,
and in human neonates, brown adipose tissue (BAT) is the site of heat
production (3, 7). The role of BAT in large primates has been contro-
versial, but recent evidence utilizing PET scanning suggests that func-
tional BAT can be demonstrated in adult humans (8, 9). The potential
role for BAT aside, it is well established that sympathetic stimulation
increases heat production in humans, that stimulation of the sympa-
thetic nervous system (SNS) underlies dietary thermogenesis, and that
suppression of the SNS with fasting or low energy diets decreases
energy expenditure (6, 10). A robust capacity for dietary thermogenesis
could act as a buffer against weight gain; a diminished capacity, on the
other hand, would decrease the range of intakes over which energy
balance could be achieved. In the later case the more efficient metab-
olism would constitute a thermogenic handicap.
Since the SNS is the mediator of adaptive thermogenesis (supressed
with fasting, increased with overfeeding), it is reasonable to ask
whether diminished SNS activity accounts for the enhanced metabolic
efficiency in the obese. This is a plausible explanation since inherited
models of rodent obesity do, in fact, have diminished sympathetic
activity (11, 12). Overfed models of obesity in rodents, however, as well
as obese humans have increased rather than diminished sympathetic
activity (13–16). Any impairment, therefore, in dietary thermogenesis is
not due to diminished SNS activity. It is possible, however that the
thermogenic response to SNS stimulation might be blunted in the
obese. This would be consistent with data suggesting that dietary
thermogenesis may be diminished in the obese.
Basal or Resting Metabolic Rate. The term resting metabolic
rate is usually employed because it assumes less about the “basal” state
than basal metabolic rate (BMR), although for practical purposes the two
290 LEWIS LANDSBERG ET AL

terms are generally considered equivalent. As noted above, in completely

sedentary man, resting metabolic rate (RMR) is considered to account for
about 80% of total energy expenditure (2). In the physically active, RMR
may be substantially less depending on the degree of activity.
RMR (or BMR), in turn, is comprised of two components: essential
energy expenditure and the energy required for homeothermy. The
former refers to the energy utilized in maintaining cellular structure and
function, including protein turnover, ionic gradients (calcium, sodium)
and the like. The latter is the metabolic cost of maintaining core temper-
ature at approximately 37 degrees C. It has been estimated that the cost
of homeothermy is about 2/3rds of the BMR (2). This is consonant with
the well recognized observation noted above that the metabolic rate of a
mouse is 8 to 10 times greater than that of a lizard of the same body
weight (3); since the components of essential energy expenditure should
be similar in both poikilotherms and homeotherms, the clear implication
is that maintaining the higher temperature is the cause of the increased
metabolic rate. If the RMR is 80% of total energy expenditure, then
approximately 50% of total energy is expended in the maintenance of
homeothermy. RMR is generated by mitochondrial metabolism regulated
by thyroid hormones with a minor contribution from the SNS.

Metabolic Rate in the Obese

Despite the compelling evidence for a thermogenic handicap in the
obese, a clear demonstration of weight-related differences in metabolic
efficiency has not been forthcoming. A likely explanation for the failure
to definitively demonstrate differences in energy expenditure in direct
comparisons between lean and obese involves factoring for body size.
Energy expenditure is measured by oxygen consumption, and this
requires normalizing for the mass of metabolizing tissue. This conun-
drum-how to compare a 150 kg person with a 70 kg person-has never
been successfully resolved. Factoring by body surface area or by lean
body mass (the usual method) has theoretical as well as practical
limitations (17). The measurement error in lean body mass determina-
tion, for example, is large enough to swamp a physiologically impor-
tant difference in oxygen consumption. Core temperature, on the other
hand, can be measured precisely and is independent of body size,
thereby avoiding this conundrum.

The Case for Core Temperature

Five points indicate the potential importance of lower core tempera-
ture as a factor in the enhanced energy efficiency associated with obesity:
 DO THE OBESE HAVE LOWER TEMPERATURES? 291

Substantial Contribution to RMR. Maintenance of a core tem-

perature of 37 degrees C is the major factor in generating an RMR. As
much as 50% of total energy expenditure might be due to homeothermy
in truly sedentary humans.
Effect of Temperature on Metabolic Rate. It has been known
since early in the 20th century that a rise in temperature is associated
with an increase in metabolic rate. Each degree C rise in temperature is
associated with a 10 –13% increment in oxygen consumption (18). The
elevation in temperature itself is responsible for speeding up metabo-
lism.
Metabolic Rate and Core Temperature Vary Among Individu-
als and in Different Populations. There is an inverse relation-
ship between mean annual ambient temperature and RMR in different
geographic regions. RMR is higher in arctic climes and lower in the
tropics (19 –22). RMR, furthermore, is known to vary among individuals
by as much as 600 Kcal per day (3).
Fall in Temperature is a Physiologic Adaptation to Decreased
Energy Intake. Given the quantitative significance of tempera-
ture maintenance in overall energy expenditure, it should come as no
surprise that a fall in temperature is part of the adaptive response to
energy deprivation. During starvation both energy output and temper-
ature drop (10, 23). Hypoglycemia, an acute state of energy deprivation,
is likewise associated with hypothermia (24 –26). Recent studies also
demonstrate that Fibroblast Growth Factor 21, a ligand generated in
the liver under the regulation of peroxisome proliferator-activated
receptor ␣ (PPAR␣), mediates lipid mobilization during fasting and
induces a fall in temperature (shallow torpor) in mice (27). All of there
observations indicate that lower temperature is an important part of
the metabolic response to diminished energy intake.
Precedence in the Animal World. Genetically obese rodents
(ob/ob mice, fa/fa rats) have lower temperatures than lean control
animals (28 –31). Hibernation, and the lesser state of shallow torpor,
which are utilized by many mammalian species, are associated with
substantial falls in core temperature (32, 33). Some indigenous human
populations also undergo torpor, an exaggeration of the normal fall in
core temperature during nighttime sleep (34).

Potential Impact of Temperature on Metabolic Efficiency

Temperature might be involved in enhancement of metabolic effi-
ciency in the following ways: lower basal core temperature set point;
greater nocturnal temperature fall; lesser temperature rise during
292 LEWIS LANDSBERG ET AL

exercise; lesser post-prandial temperature rise; and a greater temper-

ature fall during fasting or decreased energy intake as in therapeutic
dieting. Some studies in humans have suggested a lower body temper-
ature in the obese but others have not (35–37).
Quantitative Significance. Some rough estimates of the impact
of temperature on energy balance are of interest. A positive caloric
balance of 3500 – 4000 Kcal results in the deposition of one pound of fat.
Walking one mile burns about 100 kcal. A one degree C increase in core
temperature, by increasing metabolic rate 10 –13%, would lead to an
increase in caloric expenditure of 100 to 130 Kcal/day, assuming a
caloric intake of 2000 Kcal/day, and assuming that temperature ac-
counts for 50% of total energy production. This would increase the
range of dietary intake over which an individual could maintain energy
balance by a corresponding amount. Conversely, one degree C lower
temperature would impose a thermogenic handicap of 100 –130 Kcal/
day which might amount to 3000 – 4000 kcal per month or one pound of
fat. These calculations are gross approximations only, but they serve to
indicate that lower temperatures might contribute significantly to
metabolic efficiency. Temperature may thus be a component of a
thrifty metabolic phenotype.

Summary: Core Temperature and the Pathogenesis of

Obesity
“Gluttony and sloth”, over eating and lack of exercise, obviously play
an important role in the genesis of obesity. But the decisive demon-
stration that individuals differ in metabolic efficiency indicates that
other factors are also at play. Evidence summarized here suggests that
lower core temperature in the obese may be one such factor. Studies of
body temperature in the obese under different circumstances (exercise,
sleeping, after meals) are feasible and may reveal important differ-
ences between lean and obese. If this proves to be the case, a new
therapeutic target, body temperature, may emerge.

REFERENCES
1. Landsberg L, Young JB. Autonomic regulation of thermogenesis. In: Girardier L,
Stock MJ, eds. Mammalian Thermogenesis. 0 ed. London: Chapman and Hall;
1983:99 –140.
2. Girardier L, Stock MJ. Mammalian thermogenesis: an introduction. In: Girardier L,
Stock MJ, eds. Mammalian Thermogenesis. 0 ed. London: Chapman and Hall;
1983:1– 8.
3. Silva JE. Thermogenic mechanisms and their hormonal regulation. Physiol Rev
2006;86(2):435– 64.
 DO THE OBESE HAVE LOWER TEMPERATURES? 293
4. Bouchard C, Tremblay A, Despres JP, et al. The response to long-term overfeeding
in identical twins. N Engl J Med 1990;322(21):1477– 82.
5. World Health Organization. Obesity and overweight: World Health Organization;
2006. Report No. 311.
6. Landsberg L, Saville ME, Young JB. Sympathoadrenal system and regulation of
thermogenesis. Am J Physiol 1984;247:E181–9.
7. Trayhurn P, James WPT. Thermogenesis and obesity. In: Girardier L, Stock MJ, eds.
Mammalian Thermogenesis. 0 ed. London: Chapman and Hall; 1983:234 –58.
8. Nedergaard J, Bengtsson T, Cannon B. Unexpected evidence for active brown adi-
pose tissue in adult humans. American Journal of Physiology 2007;293(2):E444 –52.
9. Soderlund V LS, Jacobsson H. Reduction of FDG uptake in brown adipose tissue in
clinical patients by a single dose of propranolol. Eur J Nucl Med Mol Imaging
2006(34):1018 –22.
10. Young JB, Landsberg L. Suppression of sympathetic nervous system during fasting.
Science 1977;196(4297):1473–5.
11. Knehans AW, Romsos DR. Reduced norepinephrine turnover in brown adipose tis-
sue of ob/ob mice. Am J Physiol 1982;242:E253– 61.
12. Young JB, Landsberg L. Diminished sympathetic nervous system activity in genet-
ically obese (ob/ob) mice. Am J Physiol 1983;245:E148 –54.
13. Troisi RJ, Weiss ST, Parker DR, Sparrow D, Young JB, Landsberg L. Relation of
obesity and diet to sympathetic nervous system activity. Hypertension 1991;17:669 –
77.
14. Scherrer U, Randin D, Tappy L, Vollenweider P, Jéquier E, Nicod P. Body fat and
sympathetic nerve activity in healthy subjects. Circulation 1994;89:2634 – 40.
15. Grassi G, Seravalle G, Cattaneo BM, et al. Sympathetic activation in obese normo-
tensive subjects. Hypertension 1995;25:560 –3.
16. Vollenweider P, Tappy L, Randin D, et al. Differential effects of hyperinsulinemia
and carbohydrate metabolism on sympathetic nerve activity and muscle blood flow
in humans. J Clin Invest 1993;92:147–54.
17. Consalazio CT JR, Pecora LJ. Physiological measurements of metabolic functions in
man. New York: McGraw Hill 1963.
18. Du Bois EF. The basal metabolism in fever. Journal of the American Medical Asso-
ciation 1921;77(5):352–5.
19. Roberts DF. Basal metabolism, race, and climate. J Royal Anthropol Inst
1952;82:169 – 83.
20. Leonard WR, Sorensen MV, Galloway VA, et al. Climatic influences on basal met-
abolic rates among circumpolar populations. Am J Hum Biol 2002;14:609 –20.
21. Leonard WR, Snodgrass JJ, Sorensen MV. Metabolic adaptation in indigenous Si-
berian populations. Annu Rev Anthropol 2005;34:451–71.
22. Rode A, Shephard RJ. Basal metabolic rate of Inuit. Am J Hum Biol 1995;7:723–9.
23. Keys A, University of Minnesota. Laboratory of Physiological Hygiene. The biology
of human starvation. Minneapolis: Univ. of Minnesota Press; 1950.
24. Young JB, Landsberg L. Sympathoadrenal activity in fasting pregnant rats: disso-
ciation of adrenal medullary and sympathetic nervous system responses. J Clin
Invest 1979;64:109 –16.
25. Rappaport EB, Young JB, Landsberg L. Effects of 2-deoxy-D-glucose on the cardiac
sympathetic nerves and the adrenal medulla in the rat: further evidence for a
dissociation of sympathetic nervous system and adrenal medullary responses. En-
docrinology 1982;110:650 – 6.
26. Freinkel N, Metzger BE, Harris E, Robinson S, Mager M. The hypothermia of
294 LEWIS LANDSBERG ET AL

hypoglycemia. Studies with 2-deoxy-D-glucose in normal human subjects and mice.

N Engl J Med 1972;287:841–5.
27. Inagaki T DP, Xhao G, Ding X, Gautron L, Parameswara V, Li Y, Goetz R, Maha-
mmadi M, Esser V, Elmquest JK, Gerard RD, Burgess SC, Hammer RE, Mangels-
dorf DJ, Kliewer SA. Endocrine Regulation of the Fasting Response by PPAR␣-
Mediated Induction of Fibroblast Growth Factor 21. Cell Metabolism 2007;(5):415–
25.
28. Davis TRA, Mayer J. Imperfect homeothermia in the hereditary obese-hyperglyce-
mic syndrome of mice. Am J Physiol 1954;177:222– 6.
29. Trayhurn P, James WPT. Thermoregulation and non-shivering thermogenesis in the
genetically obese (ob/ob) mouse. Pflugers Arch 1978;373:189 –93.
30. Rising R, Keys A, Ravussin E, Bogardus C. Concomitant interindividual variation in
body temperature and metabolic rate. Am J Physiol 1992;263(4 Pt 1):E730 – 4.
31. Klaus S MH, Truloff C, Heldmaier G. Physiology of transgenic mice with brown fat
ablation: obesity is due to lowered body temperature. Am J Physiol 1998(274):R287–
R93.
32. Heldmaier G, Ortmann S, Elvert R. Natural hypometabolism during hibernation
and daily torpor in mammals. Respir Physiol Neurobiol 2004;141(3):317–29.
33. Berger RJ. Slow wave sleep, shallow torpor and hibernation: homologous states of
diminished metabolism and body temperature. Biol Psychol 1984;19(3– 4):305–26.
34. Kreider MB, Buskirk ER, Bass DE. Oxygen consumption and body temperatures
during the night. J Appl Physiol 1958;12(3):361– 6.
35. Adam K. Human body temperature is inversely correlated with body mass. Eur
J Appl Physiol 1989(58):471–5.
36. Kim H, Richardson C, Roberts J, Gren L, Lyon JL. Cold hands, warm heart. Lancet
1998;351(9114):1492.
37. Rising R, Fontvieille AM, Larson DE, Spraul M, Bogardus C, Ravussin E. Racial
difference in body core temperature between Pima Indian and Caucasian men. Int J
Obes 1995;19:1–5.

DISCUSSION
Luke, Cincinnati: Pheochromocytoses and thyrotoxicosis cause higher tempera-
tures. Is this related to this?
Landsberg, Chicago: Well both of those have a significant increase in metabolic rate
by different mechanisms, frequently by core temperature, but it’s not exclusively by core
temperature, because they sweat a lot; and so they defend their temperature but have a
lot of increased energy production.
Chapman, Jackson: Body temperatures tend to be lower in patients who are in
renal failure, uremic. Have there been any studies on weight gain that occurs in uremia?
Landsberg, Chicago: That’s a good point. I am not aware of any. Maybe some of the
nephrologists know.
Boyer, New Haven: Interesting talk, Leu. There is some recent exciting data from
France that indicates that bile acids affect thermogenesis, I think through some way of
regulating the thyroid. This would explain, perhaps, the fasting-feeding effects, since in
fasting, bile acids will stay in the gallbladder rather than being circulated; and it might
also explain the complications from ursodeoxycholic acid therapy, where weight gain is
one of the hydrophobicity of bile being decreased. Could you comment on that possibility?
Landsberg, Chicago: That is an interesting thing to pursue. We don’t know what
underlies these changes. I suspect that thyroid hormone is importantly involved and
 DO THE OBESE HAVE LOWER TEMPERATURES? 295
that thyroid-regulated mitochondrial metabolism in cells throughout the body is apt to
be important.
Boyer, New Haven: Perhaps alterations in the enterohepatic circulation of bile acid
might be playing a role here via a target for therapy.
Hochberg, Baltimore: So one of the populations which has this metabolic efficiency
and has been extensively studied are the Pima Indians in the southwest, particularly in
Arizona, and there are lots of data through NIDDK from their research station. Do you
know if anybody has ever looked at body temperature in this group?
Landsberg, Chicago: This is the only group that has been looked at in this way, and
they had two reports on temperature in the obese Pimas. One reported that it was
increased and one reported that it was decreased. It depends on how you look at it. So I
think we need further studies.